444 F. E. Conner Reed — Blind Trilohites. 



phylogenetic and ontogenetic principles deduced from recent in- 

 vestigations, is here employed. The presence of free cheeks and 

 compound eyes and the position of the facial sutures are accorded 

 the great phylogenetic importance which they are now ascertained 

 to possess. 



(1) Okder Htpoparia. 



In the order Hypoparia, which Beecher considers the lowest of the 

 three into which the Trilobita may be divided, the three families 

 Agnostidee, Harpedid^, and Trinucleidae, and the genera which they 

 include, show a combination of features characteristic of the earliest 

 larval stages of higher families. Particularly is this fact apparent 

 in the head-structure. Compound paired eyes are absent ; the 

 free cheeks form a continuous marginal ventral plate, and do not 

 show- on the upper surface of the head ; the suture is ventral, 

 marginal, or submarginal. Secondary specialization has, however, 

 in some respects obscured these primitive features and destroyed the 

 simplicity of their appearance. But in spite of these superinduced 

 characters, such as the marginal fringe in Trinucleus, and the loss of 

 segmentation in the glabella of some Agnosti, we cannot fail to 

 perceive the general stamp of a low stage of development. Of the 

 family Agnostidee, the two genera Agnostus and Microdisciis, which 

 constitute it, are blind. Matthew^ has remarked that the genus 

 Agnostus is the simplest known trilobite and shows the most perfect 

 retention of embryonic features, and that it is especially characteristic 

 of the early Cambrian period in America and also Europe, as the 

 writings of Tullberg, Brogger, and Hicks have shown. Its 

 geological distribution is thus just what we should expect from 

 its morphological features. 



In the family Harpedid^e eye-spots are frequently present on the 

 ■fixed cheeks. It has been thought that eye-spots are a primitive 

 character and of some phylogenetic importance ; but from the fact 

 that they are only found in this family, which abounds in marks 

 of secondary specialization, and that it has not been proved that 

 they are present even in any of the larval stages of other genera 

 (except in Trinucleus, which is discussed below), it seems to me that 

 we should regard them as special products of the extremely high 

 degree of secondary development reached by the members of this 

 aberrant family. 



These eye-spots may reach a considerable degree of complexity 

 in their structure, as, for instance, in Harpes maerocepJialus (Goldf.), 

 from the Devonian,- but they are in no way structures homologous 

 to the compound eyes on the free cheeks of higher trilobites. The 

 eye-spots are situated on the fixed cheeks at the end or along the 

 course of the ocular ridge or eye-line which has been thought to 

 represent the course of the optic nerve. 



A structure with a similar but more extended course, and also 

 termed the eye-line, is found in the later larval stages of some 

 primitive Olenidse (Ptychoparia, Solenopleura, etc.), and in these 



1 Trans. Eoy. Soc. Canada, vol. v (1887), p. 160. 



2 Whidborne, Monogr. Devon. Fauna, vol. i, p. 32: Palseont. Soc, 1879. 



